For three decades, Depo-Provera occupied a quiet and reassuring place in reproductive medicine: one injection, four months of contraception, nothing to remember and nothing to swallow. That quarterly simplicity is precisely what made the drug attractive to tens of millions of women. It is also, we now understand, what allowed a steady pharmacologic signal to accumulate in a tissue that listens closely to hormones. The signal is intracranial meningioma, and the question it raises is no longer whether anyone should worry, but how the law should handle a risk that science took thirty years to read.

The contraceptive at the center of this story is depot medroxyprogesterone acetate, or DMPA, a synthetic progestin delivered as a long-acting depot. It is marketed in the United States in two FDA-approved forms: Depo-Provera CI at 150 milligrams given intramuscularly, and Depo-subQ Provera 104 at 104 milligrams given subcutaneously, both dosed roughly every twelve to thirteen weeks. The molecule is a progesterone receptor agonist that crosses the blood-brain barrier. Hold that last fact in mind, because it is where the pharmacology and the pathology meet.

I.What a Meningioma Actually Is

A meningioma is a tumor that grows from the meningothelial cells of the arachnoid layer, the delicate membrane that wraps the brain and spinal cord. It is the most common primary tumor of the central nervous system, and the overwhelming majority are histologically benign, classified as WHO grade 1. Benign, however, is a word that does a great deal of quiet work. A meningioma does not metastasize, but it occupies space inside a closed skull, and space inside the skull is the one luxury the brain cannot spare. As the tumor enlarges it presses on cortex, cranial nerves, and vasculature, producing headaches, seizures, visual loss, weakness, cognitive change, and, when it sits in the wrong place, the need for craniotomy.

Two features of meningioma matter enormously for the present discussion. First, the tumor is markedly more common in women than in men, a sex disparity that has long pointed investigators toward sex hormones. Second, a substantial majority of meningiomas express the progesterone receptor on their cell surface. A tumor that carries progesterone receptors is, by definition, a tumor equipped to respond to progesterone and to synthetic progestins that mimic it. The biology was therefore primed for exactly the kind of association the epidemiology would eventually find.

A benign tumor in a closed skull is a contradiction the brain cannot afford. Space is the one luxury it does not have.

II.The Epidemiology Catches Up

The modern reckoning began in France. In March 2024, the EPI-PHARE research group, drawing on the French national health data system, published a large case-control study in The BMJ examining progestogens and intracranial meningioma. For injectable medroxyprogesterone acetate used for a year or longer, the investigators reported an adjusted odds ratio of 5.55, with a 95% confidence interval of 2.27 to 13.56, for meningioma serious enough to require surgery. Stated plainly, women with prolonged exposure were on the order of five to six times more likely to undergo an operation for an intracranial meningioma than comparable women without that exposure.

A single study, however striking, is a hypothesis with good manners, not a verdict. What gives the Depo-Provera signal its weight is that the finding did not stay in France. A 2024 analysis published in Cancers, built on a large United States insurance database of more than 117,000 meningioma cases and over a million controls, found no association with oral medroxyprogesterone acetate but roughly a 53% increase in the odds of intracranial meningioma among women exposed to the injectable form. The contrast between the oral and injectable routes is itself informative: it points toward sustained, depot-level exposure rather than the molecule alone.

The convergence continued. In June 2025, a study by Frey and colleagues in Expert Opinion on Drug Safety reported that more than a year of DMPA use carried roughly a 3.5-fold increased risk of intracranial meningioma when measured against a combined oral contraceptive comparator. Then, in September 2025, investigators from the Cleveland Clinic and Case Western Reserve published in JAMA Neurology an analysis of records spanning more than ten million women in the United States, finding an approximately 2.45-fold increased risk, with the hazard concentrated among longer-duration users and women who began the drug at older ages.

The reported risk figures range from roughly 1.5-fold to 5.6-fold across studies, populations, and designs, and that spread is not a weakness to be apologized for. It is the ordinary texture of observational epidemiology, where comparator groups, exposure definitions, and case ascertainment differ from one dataset to the next. What matters is the direction and the consistency: independent investigators on two continents, using different databases and different methods, keep finding the same arrow pointing the same way, and they keep finding it stronger with longer exposure.

III.Why the Biology Cooperates

A causal claim earns its keep when the mechanism is not merely possible but coherent with everything else that is known, and here the mechanism is unusually tidy. Medroxyprogesterone acetate is a progestin that binds and activates the progesterone receptor. Meningiomas express that receptor in the large majority of cases. And MPA, unlike some larger molecules, crosses the blood-brain barrier and reaches the intracranial compartment where these tumors live. A depot formulation, by design, holds the tissue under continuous hormonal influence for months at a stretch rather than the rising and falling exposure of a daily pill.

Put those facts together and the picture is one of sustained receptor stimulation in a hormone-responsive tissue. That is a textbook setting for promotion of growth in cells that already carry the relevant receptor. It also explains two features of the epidemiology that might otherwise look like noise: the duration-dependence, because longer stimulation means more opportunity for clinically meaningful growth, and the route-dependence, because the injectable depot sustains exposure in a way the oral route does not.

The same property that makes the depot convenient, sustained release over months, is the property that keeps a hormone-sensitive tumor under continuous instruction.

IV.The Regulatory Turn

Regulators followed the evidence, if not quickly then at least unmistakably. In December 2025, the FDA approved updated United States prescribing information for both injectable formulations, Depo-Provera CI and Depo-subQ Provera 104, adding language warning of a potential increased risk of meningioma associated with prolonged use. European regulators had already moved on the progestin-meningioma question, and the French signal had circulated in the medical literature for well over a year before the American label changed.

A label change is a small document with large consequences. For clinicians and patients, it reframes the risk-benefit conversation, particularly for long-term users and for women who began the injection at an older age. For the legal system, the timing of that warning becomes a fact of central importance. Every woman injected before December 2025 received a product whose American labeling said nothing about this risk, and the gap between when the manufacturer knew or should have known and when the warning appeared is the territory on which failure-to-warn litigation is fought.

V.From Pharmacology to the Courtroom

The litigation has grown quickly. Federal cases have been consolidated into multidistrict litigation, MDL No. 3140, In re: Depo-Provera (Depot Medroxyprogesterone Acetate) Products Liability Litigation, before Judge M. Casey Rodgers in the United States District Court for the Northern District of Florida. From a few hundred filings in mid-2025, the docket has climbed into the thousands, with claims alleging that Pfizer failed to adequately warn patients and physicians of the meningioma risk and that the product was defectively designed. The cases now sit at the procedural inflection point that decides most pharmaceutical mass torts: general-causation expert challenges. Daubert hearings on whether the science is admissible are calendared for late June 2026, with a first bellwether trial anticipated at the end of that year, and a manufacturer preemption motion pending in parallel.

This is where forensic pharmacology stops being background and becomes the case. Mass-tort causation is litigated on two levels, and they must not be confused. General causation asks whether DMPA is capable of causing intracranial meningioma in human beings at all. That is the question the four studies above are marshaled to answer, and it is the question the Daubert gatekeeping will test. Specific causation asks whether this drug caused this tumor in this plaintiff, and it is decided one woman at a time on the particulars of her exposure and her disease.

The honest expert says the quiet part out loud: the absolute risk to any single woman remains small. A relative risk of two-fold or five-fold operates on a baseline incidence that is itself low, and one study placed the number needed to harm for the injectable at roughly one additional meningioma per thousand-plus users. This candor is not a concession that weakens a claimant’s case. It is what makes the rest of the testimony believable, and it is the difference between an opinion that survives cross-examination and one that does not. A risk can be real, biologically coherent, replicated across populations, and serious for the affected individual, while still being statistically uncommon. All of those things are true here at once.

The classic Bradford Hill considerations sit comfortably over this evidence. There is strength of association in the higher estimates, consistency across independent datasets and countries, a biological gradient in the duration-dependence, biological plausibility in the receptor pharmacology, and coherence between the laboratory picture and the population data. None of these is a checkbox that decides causation by itself, and a competent expert resists the temptation to treat them as one. Together, applied with judgment rather than as a tally, they describe a body of evidence that a court is entitled to hear.

VI.What It Means Going Forward

For the clinician, the practical lesson is proportion rather than alarm. DMPA remains an effective contraceptive, and for many patients the benefit still outweighs a small absolute risk. The conversation that the new label invites is a frank one about duration, about reassessing very long-term use, and about remaining alert to new headaches, visual changes, or seizures in a long-term user, symptoms that warrant imaging rather than reassurance.

For the lawyer and the forensic scientist, the lesson is that this is a case built on documentable facts, not on rhetoric. Exposure records, imaging, surgical pathology, and receptor status are the load-bearing elements, and the strength of any individual matter rises or falls on how cleanly those facts line up with a coherent mechanism and a defensible latency. The science here is good enough to take seriously and specific enough to test. That combination, evidence that is both credible and falsifiable, is exactly the ground on which causation should be decided.

Thirty years passed between the drug’s introduction and the warning that now accompanies it. The interval is a reminder that pharmacovigilance is slow, that depot drugs accumulate their signals quietly, and that the tissues most sensitive to a molecule are not always the ones we are watching. The needle was simple. The meninges were listening. The law is now asked to reconcile the two.